Bronchoalveolar lavage in asthmaticchildren: Evidence of neutrophil activation inmild-to-moderate persistent asthma

Barbato A, Panizzolo C, Gheno M, Sainati L, Favero E, Faggian D,Giusti F, Pesscolderungg L, La Rosa M. Bronchoalveolar lavage inasthmatic children: Evidence of neutrophil activation in mild-to-moderate persistent asthma.Pediatr Allergy Immunol 2001: 12: 73–77. # Munksgaard, 2001

Little information is available on cell profiles and mediator production inthe lower airways of children with asthma by comparison with the adultpopulation. To study the bronchoalveolar lavage (BAL) cell profiles andproduction of eosinophil cationic protein (ECP) and myeloperoxidase(MPO) in childhood bronchial asthma, a retrospective study wasperformed in 29 children (13 allergic asthmatic children and 16 controls).Six of the asthmatics had mild-to-moderate persistent disease and sevenhad intermittent asthma. The BAL cell count and ECP and MPO values ofasthmatic children were compared with those from 16 controls. Theasthmatic patients had higher values than controls for the total cell count(p50.08), for neutrophils (p50.02), and for ECP and MPO (p,0.001).MPO levels (p50.04), neutrophil count (p50.06), and ECP values(p50.06) were higher in patients with mild-to-moderate persistent asthmathan in those with intermittent asthma. Our results demonstrate thatneutrophil-mediated inflammation is greater in patients with more severeasthma.

Angelo Barbato1, Cristina Panizzolo1,Marta Gheno1, Laura Sainati1,Elisabetta Favero2, Diego Faggian3,Franca Giusti4,Lidia Pesscolderungg5 andMario La Rosa6

1Department of Pediatrics, 2School of Allergy,3Department of Laboratory Medicine and4Department of Anesthesiology, University ofPadova, Padova, Italy, 5Pediatric Division,Bolzano Hospital, Bolzano, Italy, 6Departmentof Pediatrics, University of Catania, Catania,Italy

Key words: BAL; asthmatic children;

neutrophils; MPO; pediatric

Prof. Angelo Barbato, Department ofPediatrics, Via Giustiniani 3, 35128 Padova, Italy

Tel.:+39-049-8213505

Fax: +39-049-8213502E-mail: barbato@child.pedi.unipd.it

Accepted 30 November 2000

Bronchial asthma in childhood is a disease thatoften starts in children when under 3 years ofage and persists as they grow up (1). Itsprevalence is increasing (2) and therefore abetter understanding is needed for a moreadequate diagnosis and treatment, especiallyin younger children.

New research opportunities have becomeavailable with the introduction of fiberopticbronchoscopy (FB) and bronchoalveolar lavage(BAL), which are being used to an increasingextent in pediatric pneumology centres (3).

There are reports of high levels of eosino-phils in the BAL collected from asthmaticchildren (4). Little is known, however, aboutthe role of neutrophils in the chronic inflam-mation induced by bronchial asthma, describedin recent studies on adults (5,6) and children(7), as cells that remain activated in asthmadisease.

The aim of the present study was to contributetowards a better understanding of this aspect ofasthmatic inflammation.

Subjects and methods

Subjects

As, for ethical reasons, we could not study the cellprofiles and mediators in the BAL fluid of childrenwith asthma, we performed a retrospective study inallergic children with asthma (8) who had under-gone FB and BAL for other indications (3).

The study included 13 asthmatic children (sevenmales, six females; mean age 7 years 8 months,range 3 years 3 months to 17 years 7 months) whohad been hospitalized in the Department ofPediatrics, Padova from December 1997 to May1999 to undergo FB for the following indications:eight for recurrent chest radiographic shadowing

Pediatr Allergy Immunol 2001: 12: 73–77

Printed in UK. All rights reservedCopyright # Munksgaard 2001

PEDIATRIC ALLERGYAND IMMUNOLOGY

ISSN 0905-6157

73

in the same lobe (four in the middle lobe, three inthe lower left lobe, and one in the upper left lobe) –which dated from an average of 3 months (range1–6 months) before FB and BAL – two for per-sistent atelectasis, and three for chronic cough. Thediagnosis of asthma was based on the NationalHeart Lung and Blood Institute (NHLBI) andWorld Heath Organization (WHO) guidelines (9).Seven children suffered from intermittent asthma,reporting symptoms less than twice a week, night-time symptoms less than twice a month, andcompletely symptom-free periods between exacer-bations. The symptoms were controlled by inhaledbronchodilators. Six children suffered from mild-to-moderate persistent asthma with symptomsoccuring more than twice a week and/or once aweek at night, and needed continuous inhaledsteroid administration (the average dose of beclo-methasone dipropionate was 586 mg/day, range500–750 mg) up until the day when FB and BALwere performed.

All children were skin-prick test or radio-allergosorbent test (RAST) positive for inhaledallergens.

Nine children underwent spirometry (flow–volume curves were measured by using thePulmonet-III, Sensor Medics, CA, USA)(Table 1), which documented varying degrees ofairflow obstruction (that improved both sponta-neously and after beta-2 inhalation).

All children were immunocompetent and nonehad laboratory signs of acute infection or werereceiving antibiotic treatment at the time ofhospitalization.

The control group comprised 16 children (ninefemales and seven males; mean age 8 years10 months, range 3 years 6 months to 16 years3 months), who underwent bronchoscopic BALduring elective surgery for non-pulmonary dis-ease (10 for adenoidectomy and/or tonsillectomy,six for minor surgery). As medical examinationprior to surgery revealed no clinical history ofasthma or atopic disease, these children under-went no allergy tests. Informed parental consentwas obtained for all children taking part in thestudy.

FB and BAL

The FB and BAL were performed by the sameoperator using an Olympus 3C30 or P10fiberoptic bronchoscope, depending on whetherthe children were younger or older than 9 years ofage. During the procedure, patients were sedatedwith intravenous (i.v.) propofol (3 mg/kg) andatropine was used for premedication. Localanesthesia was induced in the lower airways

with lidocaine 0.5%. BAL was performed in thelung segment that, judging from endoscopicexamination and the most recent chest X-ray,appeared to be most greatly affected by theinflammatory process; failing this indication,BAL was performed in the middle lobe. Theamount of saline solution injected for each childwas 1 ml/kg of body weight and was divided intothree aliquots of 5–20 ml, depending on theweight of the patient (10): the first fraction wasused for culture and the other two were pooled(11) and sent to the laboratory for analysis oftotal and differential cell counts, and for assay ofmyeloperoxidase (MPO) and eosinophil cationicprotein (ECP) levels.

Cytological analysis and mediator assay

Before assessing the cellular and non-cellularcomponents, the fluid was centrifuged at 303gfor 10 min. The supernatant was used for assayof MPO and ECP levels by radioimmunoassay(Pharmacia-Upjohn, Uppsala, Sweden); resultswere expressed in mg/l. The lower limit ofdetection for ECP was 1 mg/l and for MPO itwas 40 mg/l.

After eliminating the supernatant, the cellpellet was suspended in 1 ml of Hank’s balancedsalt solution (Ca2+- and Mg2+-free) and centri-fuged (Phicom Co., Cheshire, UK), spinning< 13106 cells/ml. The centrifuged cell prepara-tions were air-dried and stained with MayGrunwald-Giemsa stain.

Cell evaluation was performed by the sameinvestigator, blinded to the patients’ clinicaldetails, using a light microscope (Leitz-Dialux2,Ernst Leitz, Weitzlar, Germany). At least 1000cells were examined and 100 were counted foreach patient. A small sample of cell suspensionwas used for evaluation of total cell counts, usingthe Coulter-Z1 hemocytometer.

The total cell values and the different cellpopulations were expressed as 3103/ml.

Statistical analysis

Age was expressed as mean and range, forcedexpiratory volume in 1 s (FEV1) as a percentageof the reference values, and total immunoglobulinE (IgE) as median and range.

The amount of BAL fluid obtained (expressedas a percentage of the instilled saline solution) ispresented as mean6SD. The total cell anddifferential cell counts are expressed per ml offluid recovered and shown as medians and ranges.ECP and MPO are expressed as medians andranges. The non-parametric Mann–WhitneyRank Sum test was used to compare results of

Barbato et al.

74

asthmatics and controls. The same test was usedto compare the intra-group results betweenpatients with intermittent or persistent asthma.

The correlation between the number of neu-trophils and MPO was obtained using Pearson’scorrelation coefficient.

A probability value of ,0.05 was consideredstatistically significant.

Results

The total and differential cell counts in theBAL fluid from the asthmatic children andcontrols are shown in Table 2.

Total cell counts were much higher in theasthmatic children (median 13143103/ml) thanin the controls (median 6253103/ml), p5 0.08.The numbers of eosinophils, lymphocytes, andepithelial cells were not significantly differentstatistically between the two groups of children,whereas the neutrophil levels were significantly

higher in the asthmatics (median 803103/ml)than in the controls (median 263103/ml),p5 0.02. ECP levels were higher in the asth-matics (median 24 mg/l) than in the controls(2 mg/l), p,0.001, as were the MPO values(median 189 mg/l in asthmatics vs. 40 mg/l incontrols, p,0.001).

When the neutrophil counts and the corre-sponding values of MPO were analyzed inasthmatic children, a linear correlation wasfound with an Rs of 0.73 (p,0.005). Nosignificant correlation was found between ECPvalues and eosinophil counts (r5–0.10;p5 0.74); a positive trend was found betweenECP values and neutrophil counts (r50.50;p5 0.07).

When the group with asthma was dividedaccording to disease severity (six patients hadmild-to-moderate persistent asthma and sevenpatients had intermittent asthma), neutrophilcounts (p5 0.06), ECP values (p5 0.06) and

Table 1. Clinical features of asthmatic patients

Subject no. Gender Age FEV1 (%) Total IgE (IU/ml) Allergy Inhaled steroid

1 M 3 y 3 m – 30 G No2 M 11 y 6 m 92 504 HDM Yes3 F 6 y 1 m 105 472 HDM Yes4 F 7 y 11 m 93 1575 HDM Yes5 M 17 y 7 m 89 1172 G No6 M 8 y 3 m 97 121 C No7 M 7 y 2 m 87 32 HDM Yes8 F 3 y 11 m – 8 B No9 F 10 y 2 m 90 252 A No

10 F 9 y 11 m 86 489 C Yes11 M 8 y 2 m 104 418 HDM Yes12 F 5 y 5 m – 94.3 A No13 F 4 y 9 m – 306 C No

7 y 8 m* 306{(3 y 3m–17 y 7 m){ (8–1575){

A, alternaria, B, betulla; C, cat; HDM, house dust mites; G, grasses.FEV1, forced expiratory volume in 1 s; IgE, immunoglobulin E; m, months; y, years.*Mean.{Range.{Median.

Table 2. Cell population and eosinophil cationic protein (ECP) and myeloperoxidase (MPO) values in bronchoalveolar lavage (BAL) of asthmatic children and controls

Asthmatics Controls p-value

No. of patients 13 16BAL recovered (%) 43610.8 37.7610.7 0.32Total cell count* (3103/ml) 1314 (84–4000) 625 (144–1167) 0.08Alveolar macrophages* (3103/ml) 714 (51–1890) 542 (225–2088) 0.79Neutophils* (3103/ml) 80 (12–1191) 26 (0–144) 0.02Eosinophils* (3103/ml) 0 (0–800) 0 (0–42) 0.13Lymphocytes* (3103/ml) 75 (0–960) 21 (0–84) 0.08Epithelial cells* (3103/ml) 5 (0–280) 6 (0–144) 0.41ECP* (mg/l) 24 (2–180) 2 (2–10) ,0.001MPO* (mg/l) 189 (40–5500) 40 (40–85) ,0.001

*Values are expressed as median and ranges (in brackets).

Neutrophil activation in BAL of asthmatic children

75

MPO values (p ,0.04) were significantly higherin the group with mild-to-moderate persistentasthma, while no significant differences wereobserved in the total number of alveolarmacrophages, eosinophils, lymphocytes, andepithelial cells (Table 3 and Fig. 1). The meanage of patients in the group with intermittentasthma did not differ significantly from thoseof the persistent asthma group (mean ages101.5 months and 91.4 months, respectively)confirming that the results obtained wereunlikely to depend on the children’s age orthe size of bronchoscope used.

In our series, the indication for FB wasrecurrent chest radiographic shadowing in samelobe in eight patients; four of these patients hadintermittent asthma and four had persistentasthma.

Six BAL cultures (four in patients withintermittent asthma and two with persistentdisease) were positive for bacterial growth. Thecultures from the four children with intermit-tent asthma were positive for Haemophilusinfluenzae (three patients) and Streptococcuspneumoniae (one patient); the two positivecultures from children with persistent asthmawere positive for S. pneumoniae in one and for

H. influenzae in the other. No differences werefound among the cell counts and mediatorvalues of these six patients and the patientswith a negative BAL culture.

No differences were found between the cellcounts and mediator values of patients withrecurrent chest radiographic shadowing in samelobe and those with persistent atelectasis andchronic cough.

Discussion

A major constraint on BAL studies in childrenwith asthma is the ethical difficulty of usinginvestigative procedures that carry some degreeof discomfort. There is consequently evidenceof substantial benefit only when a complicationof asthma justifies the use of FB and BAL. Thislimitation means that only selected asthmaticchildren can be studied using BAL.

Our asthmatic patients had higher total cellcount than controls, higher levels of neutrophilsand MPO (a mediator that expresses their acti-vation), and higher levels of ECP with no signifi-cant difference in the number of eosinophils. Thehigher total cell count in the asthmatic patientsthan in the controls is even more significant if weconsider that the total cell count observed in theBAL of our controls is higher than reportedelsewhere (11,12), although not significantly so –the slight increase might be explained by the factthat 10 of our controls underwent BAL duringadenotonsillectomy for adenotonsillar hyper-trophy with narrowing of the upper airways.

The inflammation caused by asthma explainsthe higher ECP levels. An increase in ECP andeosinophils in the BAL of asthmatic childrenafter non-specific bronchial challenge withhistamine has been reported (4). More recently,high ECP levels were also recorded in theBAL of asthmatic children without bronchialstimulation and during relatively asymptomatic

Table 3. Cell population and eosinophil cationic protein (ECP) and myeloperoxidase (MPO) values in bronchoalveolar lavage (BAL) in children with mild-to-moderatepersistent asthma (six patients) and in children with intermittent asthma (seven patients)

Persistent asthma Intermittent asthma p-value

No. of patients 6 7Total cell count (3103/ml) 1135 (571–3000) 850 (84–4000) 0.52Alveolar macrophages* (3103/ml) 588 (51–1890) 714 (67–1400) 0.77Neutophils* (3103/ml) 282 (48–1191) 58 (12–840) 0.06Eosinophils* (3103/ml) 27 (0–120) 0 (0–800) 0.62Lymphocytes* (3103/ml) 50 (0–510) 78 (0–960) 0.47Epithelial cells* (3103/ml) 51 (0–280) 5 (0–96) 0.37ECP (mg/l) 84.5 (12–180) 14 (2–28 0.06MPO (mg/l) 614 (40–5500) 46 (40–239) ,0.04

*Values are expressed as median and ranges (in brackets).

Fig. 1. Myeloperoxidase (MPO) values in six children withmild-to-moderate persistent asthma (1) and in seven childrenwith intermittent asthma (2).

Barbato et al.

76

periods (13). In our study of asthmatic childrenwithout any non-specific bronchial stimulation,we found high ECP levels with no increase inthe eosinophil count. This is consistent with arecent report (14) demonstrating in biopsyspecimens that the severity of bronchial inflam-mation is greater in persistent asthma than inintermittent asthma.

The inflammation caused by asthma couldalso explain the rise in neutrophil levels.Neutrophil activation has already been reportedin asthmatic adults (5,6) and also, morerecently, in asthmatic children, and it is relatedto symptom severity (7). In our opinion, ourfindings were not influenced by the positiveBAL culture, given that positivity was alsofound in four children with intermittent asthma(with lower neutrophil and MPO counts) and inonly two children with persistent asthma, whohad higher neutrophil and MPO levels. Thepositive trend between ECP values and neu-trophil counts observed in our patients is inagreement with the relationship previouslyreported in adults, where raised levels of ECPand neutrophils, but not eosinophils, werespecifically associated with severe, persistentdisease (15). The relationship between ECPvalues and neutrophil number could beexplained by the fact that ECP might reflecteosinophil turnover, while neutrophil phagocy-tosis of eosinophil granules could lead to anunderestimation of eosinophil counts (16).

In our asthmatic patients these results demon-strate an ongoing neutrophil inflammation that ismore relevant in children with mild-to-moderatepersistent asthma than in patients with inter-mittent asthma, while the eosinophil activity (interms of number and mediator production) doesnot seem to differ substantially between the twogroups of asthmatic patients.

In conclusion, our findings demonstrated thatmore severe asthma, requiring the use ofinhaled steroids to control the patient’s symp-toms, is correlated with a persistent, neutrophil-mediated inflammation.

References

1. MARTINEZ FD, WRIGHT AL, TAUSSIG LM, et al. Asthmaand wheezing in the first six years of life. N Engl J Med1995: 332: 133–8.

2. SLY RM. Changing prevalence of allergic rhinitis andasthma. Ann Allergy Asthma Immunol 1999: 82: 233–48.

3. BARBATO A, MAGAROTTO M, CRIVELLARO M, et al. Use ofbronchoscope, flexible and rigid, in 51 European Centres.Eur Respir J 1997: 10: 1761–6.

4. FERGUSON AC, WHITELAW M, BROWN H. Correlation ofbronchial eosinophil and mast cell activation withbronchial hyperresponsiveness in children with asthma.J Allergy Clin Immunol 1992: 90: 609–13.

5. LAMBLIN C, GOSSET P, TILLIE-LEBLOND I, et al. Bronchialneutrophilia in patients with non-infectious status asth-maticus. Am J Respir Crit Care Med 1998: 157: 394–402.

6. FRANGOVA V, SACCO O, SILVESTRI M, et al. BAL neutro-philia in asthmatic patients. A by-product of eosinophilrecruitment? Chest 1996: 110: 1236–42.

7. MARGUET C, JOUEN-BOEDES F, DEAN TP, WARNER JO.Bronchoalveolar cell profiles in children with asthma,infantile wheeze, chronic cough, or cystic fibrosis. AmJ Crit Care Med 1999: 159: 1533–40.

8. WARNER JO, NASPITZ CK, CROPP GJA. Third internationalpediatric consensus statement on the management ofchildhood asthma. Pediatr Pulmonol 1998: 25: 1–17.

9. GLOBALINITIATIVEFORASTHMA. Global Strategy for AsthmaManagement and Prevention. NHLBI/WHO WorkshopReport: Publication no. 95/3659, 1995.

10. DE BLIC J, MCKELVIE P, LE BOURGEOIS M, et al. Value ofbronchoalveolar lavage for the diagnosis of pneumonia inimmunocompromised children. Thorax 1987: 42: 759–65.

11. RIEDLER J, GRIGG J, STONE C, et al. Bronchoalveolar lavagecell count in healthy children. Am J Respir Crit Care Med1995: 152: 163–8.

12. MIDULLA F, VILLANI A, MEROLLA R, et al. Bronchoalveolarlavage studies in children without parenchymal lungdisease: cellular constituents and protein levels. PediatrPulmonol 1995: 20: 112–8.

13. ENNIS M, TURNER G, SCHOCK BC, et al. Inflammatorymediators in bronchoalveolar lavage samples fromchildren with and without asthma Clin Exp Allergy1999: 29: 362–6.

14. VIGNOLA AM, CHANEZ P, CAMPBELL AM, et al. Airwayinflammation in mild intermittent and in persistentasthma Am J Respir Crit Care Med 1998: 157: 403–9.

15. JATAKANON A, VASUF C, MAZIAK W, LIM S, CHUNG KF,BARNES PJ. Neutrophilic inflammation in severe persistentasthma. Am J Respir Crit Care Med 1999: 160: 1532–9.

16. ABU-GHAZALEH RI, DUNNETTE SL, LOEGERING DA, et al.Eosinophil granule proteins in peripheral blood granu-locytes. J Leukoc Biol 1992: 52: 611–8.

Neutrophil activation in BAL of asthmatic children

77